I think it's a good idea, especially if you could do visual aspect changes as well (custom installer, desktop environment, backgrounds, icon placement, menu's, etc.).

On Fri, 2004-03-26 at 11:20, Adam T. Gautier wrote: > I had the need to build custom installs of Fedora. So, I have developed > a series of python scripts that greatly reduce headache of building and > maintaining a custom distribution. Obviously I just use Anaconda and > it's utilities to produce the distribution. But, to me, even with lots > of scripts to automate tasks such as package dependency resolution and > comp file generation the real problem is assembling the packages from > around the Internet and starting from pre-built solutions. > > I was curious if anyone would like a service that could easily produce > custom Linux distros. I did some searching and there seems to be a > defunct company that offered for $250 + $/hr to produce a custom Linux > distro, I think that was a bit too high a price. I was thinking of a > webservice that would allow the user to choose packages from various > repositories around the Internet to build a distro-spec. This distro > spec would be public to all who wanted to start from that point and > customize their own distro. To build the distro a GUI client would be > downloaded and installed that downloads the distro spec and assembles > the packages on the local HD, customizes images, allow for custom > packages, etc... > > I was thinking that this could be a subscription based service > $10-$15/year which allows a user to download and create as many distro > specs as wanted (probably some limitation like, user can only have 5 > distro specs in development at any one time, A user would publish the > distro to the world to free up space). The pricing is just to cover the > costs of hosting and bandwidth (big bandwidth hog). My estimates are > for a break even at 25 users/month @ $10/year or 17 @$15. Anything, > above that gets pumped into development which will all be open-source > anyway. I am not thinking this is a million dollar idea because of the > small market, I would like to get some consulting out of this so I can > quit my real job :) > > > The features that I could build into the service in about 6 months are: > > *.) Website to host system > 1.) Full RPM package library to choose packages from > 2.) Distro spec rating system > 3.) Pre packaged distro specs such as: > email server distro: > windows file server distro > CMS distro > Gaming system distro > ... > 4.) expand to other distros like Mandrake, Suse and Debian > 5.) Include support for apt packages > > > If I choose to do this project it will be Fedora based in the beginning > so that is why I am posting here. I just wanted to get some feedback on > the idea, I could have this service test launched by end of April since > I already have most of the pieces developed for myself. > > So what does everyone think? -- Adam Voigt adam(a)kotisprop.com -- fedora-list mailing list fedora-list(a)redhat.com To unsubscribe: http://www.redhat.com/mailman/listinfo/fedora-list

Also, I am not thinking that the any joe schmo would use this service. It would really be for IT managers, developers, and technical hobbiest. They would choose the packages they wanted available to be installed. Instead of using the one size fits all Fedora install, which also has the contraint of only using packages allowed by RedHat, no third party packages.

That is why I want to base the service on existing distributions. Open source is just standing on the shoulders of giants. A user could throw a bunch of packages together and hope that they worked but that is their perogative, I could do that now if I wanted but I know it is a waste of time.

>that the >installer is consistent and easy to use, etc, etc, and > Agreed, I would work to make the existing installers better. >then provide some means >of ongoing maintenance (even if it's as simple as "here's a diff file, there's >the code, have fun"). > > > I agree, Red Hat, Mandrake, and Suse already have thriving businesses based on the support model. I expect users to get their support packs and package upgrades etc from them. I just think there is a group of people that want to define their own distribution parameters: look and feel, packages, configuration, etc... If these systems are based on say Fedora why can't they get support from these mailing lists. Would it not be the same as installing fedora and editing a file in /etc? >This process is not conduive to, nor indeed feasable, via a "public one size >fits all" webpage service. > Why not? > If you have something else in mind, like a >"pre-customize Fedora Core with your own logos" or something, that's another >thing entirely. You could just release your own version of fedora-logos. > > > Creating custom RPMs is one way to go... But people use technology to implement solutions not to install software. I am just trying to provide another distribution of complete solution sets not just installable software. Sure I could create a custom RPM of postfix to implement a specific part of a solution, I could also create a pre-configured distribution that I know will have the correct packages for the solution. Just seems like another way to go...

I had the need to build custom installs of Fedora. So, I have developed a series of python scripts that greatly reduce headache of building and maintaining a custom distribution. Obviously I just use Anaconda and it's utilities to produce the distribution. But, to me, even with lots of scripts to automate tasks such as package dependency resolution and comp file generation the real problem is assembling the packages from around the Internet and starting from pre-built solutions.

>Sounds a lot like what kickstart already can do + extra packages and a >web frontend. > > > Kickstart is a big part of the idea, if not only because of it's importance in relation to Anaconda. The web portion is to make package resolution easier. Which in my opinion is a pain in the but right now. The web portion would span lots of third-party repositories. That is the reason it is web-based: I think this can only be solved in a networked setting and a webapp is the fastest/easiest way to gurantee that the system is in a networked environment with updated package info. Also, kickstart can add customization through scripting only (pre and post scripts) which is fine but I also want to build a customized Anaconda that is more modular and can have some pre-built functions (Example: Configuration for apache that allows finer grained control of the modules being initialized, sure you could edit the file in /etc but if you are install 5-10 in a loadbalancing situation it becomes a pain). These customizations may best be addressed in the first boot script or Anaconda, depends on the job. Also, all of the additions to Anaconda should be easily scriptable using kickstart.

Out of curiosity, besides not being based on Fedora, what about LFS doesn't suit your needs? http://www.linuxfromscratch.org/index.html